AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Using Skills

skill-int2t05-engineering-skills-using-skills · by int2t05

Use when starting a session or deciding which skill applies — maps incoming work to the right skill across the 9 SDLC phases. Triggers on "which skill should I use", "route this task", "用哪个技能", "路由".

No reviews yet
0 installs
21 views
0.0% view→install

Install

$ agentstack add skill-int2t05-engineering-skills-using-skills

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-int2t05-engineering-skills-using-skills)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
19d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Using Skills? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Using Skills

A router. Identify the SDLC phase of the incoming work, then activate the matching skill. Every skill in this collection lives under one of nine phases.

When to use

  • Start of a session, before any response or action.
  • Unsure which skill fits the task at hand.
  • A task spans multiple phases and you need to sequence them.

Not for: mid-task when a skill is already active — don't re-route. For any new task, route even if it seems obvious; the skill may enforce discipline the shortcut skips.

Rationalizations that skip routing — resist them:

| Excuse | Reality | |---|---| | "This is just a simple question." | Questions are tasks. A skill may structure the answer. | | "I already know how to do this." | Knowing how ≠ following the discipline. The skill enforces the steps. | | "It's faster to just do it." | A skipped test-first or review is a bug shipped faster. | | "The skill is obvious here." | Obvious to route ≠ obvious to follow. Invoke it; let it run. | | "This is too small for a skill." | Small tasks are where discipline slips. tdd/debugging scale down. | | "I'll route after I start." | Starting unrouted biases you against the skill's process. Route first. |

Steps

  1. Pick the phase, then activate the entry-point skill it lists.
  • metausing-skills (this router) · skill-authoring (author + eval skills)
  • 01-productbrainstorm · spec · oss-strategy
  • 02-researchresearch (general / market / tech-selection modes)
  • 03-designarchitecture · domain-modeling · api-design · codebase-design · frontend-design · image-to-code · imagegen · design-research · schema-design · prompt-engineering · prototype
  • 04-developimplement · multi-agent-orchestration · breakdown · context-engineering · i18n · auth-implementation · error-handling
  • 05-tunecost-optimization · performance · simplify · refactoring
  • 06-testtdd · test-generation · api-testing · e2e-testing · load-testing
  • 07-verifya11y-review · code-review · debugging · security-review · linting
  • 08-shipshipping · git-workflow · ci-cd · deprecation-migration · oss-polish
  • 09-operateobservability · documentation-audit · handoff · incident-response
  1. Disambiguate with three questions:
  • Building new, or fixing what exists? New → start at 01-product (brainstorm / spec), not 04-develop. Fixing → 07-verify (debugging), then 06-test (tdd) to lock the fix.
  • Process question or implementation question? Process skills (brainstorm, debugging, code-review) run before implementation skills (frontend-design, api-design).
  • About to ship? Run 08-ship in order: git-workflowci-cdshipping.
  1. Sequence multi-phase work in phase order; finish one skill before starting the next. A full feature may run product → design → develop → test → verify → ship. A bug fix may be just debuggingtddcode-reviewgit-workflow.
  1. Skills are rigid (TDD, debugging — follow exactly) or flexible (patterns — adapt to context). The skill's own body tells you which.
  1. For the full decision tree — every branch routed to its skill — see references/phase-tree.md.

Verify

The routed skill's ## When to use matches the task, and the skill activates via the Skill tool without a mismatch signal. If the skill's triggers don't fit the work, re-route — don't force-fit.

If no skill fits after disambiguation — don't force a poor match:

  1. Say so — name the gap explicitly rather than silently improvising.
  2. Use general engineering ability to help directly (the principles are always loaded).
  3. If the gap is recurring, suggest the user invoke skill-authoring — the pack grows by

filling real needs, not speculative ones.

References

  • [${CLAUDEPLUGINROOT}/references/engineering-principles.md](${CLAUDEPLUGINROOT}/references/engineering-principles.md) — discipline shared by every skill in this collection.
  • [references/phase-tree.md](references/phase-tree.md) — full ASCII decision tree by phase and task shape.

External skill discovery (npx skills find/add) is out of scope — this router maps tasks to this collection's 47 internal skills only.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.